AI Gateway

List gateways

GET/accounts/{account_id}/ai-gateway/gateways

Get a gateway

GET/accounts/{account_id}/ai-gateway/gateways/{id}

Create a gateway

POST/accounts/{account_id}/ai-gateway/gateways

Update a gateway

PUT/accounts/{account_id}/ai-gateway/gateways/{id}

Delete a gateway

DELETE/accounts/{account_id}/ai-gateway/gateways/{id}

ModelsExpand Collapse

AIGatewayListResponse object { id, cache_invalidate_on_update, cache_ttl, 27 more }

id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

cache_invalidate_on_update: boolean

cache_ttl: number

minimum0

collect_logs: boolean

created_at: string

formatdate-time

modified_at: string

formatdate-time

rate_limiting_interval: number

minimum0

rate_limiting_limit: number

minimum0

authentication: optional boolean

Require authentication for gateway access. New gateways must have authentication enabled. Once enabled, authentication cannot be disabled.

byok_only: optional boolean

Requires customer-provided provider credentials and prevents fallback to Unified Billing.

dlp: optional object { action, enabled, profiles } or object { enabled, policies }

One of the following:

object { action, enabled, profiles }

action: "BLOCK" or "FLAG"

One of the following:

"BLOCK"

"FLAG"

enabled: boolean

profiles: array of string

object { enabled, policies }

enabled: boolean

policies: array of object { id, action, check, 2 more }

id: string

action: "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

check: array of "REQUEST" or "RESPONSE"

One of the following:

"REQUEST"

"RESPONSE"

enabled: boolean

profiles: array of string

fedramp_level: optional "fedramp-medium" or "fedramp-high"

FedRAMP compliance level: fedramp-medium or fedramp-high.

One of the following:

"fedramp-medium"

"fedramp-high"

fedramp_mode: optional "prefer" or "require"

FedRAMP enforcement mode. “prefer” prioritizes compliant connectors but allows fallback. “require” strictly filters.

One of the following:

"prefer"

"require"

guardrails: optional object { prompt, response }

prompt: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

response: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

is_default: optional boolean

log_classification: optional boolean

log_management: optional number

maximum10000000

minimum10000

log_management_strategy: optional "STOP_INSERTING" or "DELETE_OLDEST"

One of the following:

"STOP_INSERTING"

"DELETE_OLDEST"

logpush: optional boolean

logpush_public_key: optional string

maxLength1024

minLength16

otel: optional array of object { headers, url, authorization, content_type }

headers: map[string]

url: string

formaturi

maxLength2048

authorization: optional string

maxLength256

content_type: optional "json" or "protobuf"

One of the following:

"json"

"protobuf"

rate_limiting_technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

retry_backoff: optional "constant" or "linear" or "exponential"

Backoff strategy for retry delays

One of the following:

"constant"

"linear"

"exponential"

retry_delay: optional number

Delay between retry attempts in milliseconds (0-60000)

maximum60000

minimum0

retry_max_attempts: optional number

Maximum number of retry attempts for failed requests (1-5)

maximum5

minimum1

spend_limits: optional object { enabled, rules }

enabled: optional boolean

rules: optional array of object { limit, limitType, window, 6 more }

limit: number

exclusiveMinimum

minimum0

limitType: "cost"

window: number

exclusiveMinimum

minimum0

id: optional string

minLength1

enabled: optional boolean

metadata: optional map[object { mode } or object { mode, values } ]

One of the following:

Mode object { mode }

mode: "partition"

object { mode, values }

mode: "filter"

values: array of string

model: optional object { mode, values }

mode: "filter"

values: array of string

provider: optional object { mode, values }

mode: "filter"

values: array of string

technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

store_id: optional string

Secrets Store used by this gateway. Omit on update to keep the current store. Cannot be cleared while stored keys or secret references depend on it.

stripe: optional object { authorization, usage_events }

authorization: string

usage_events: array of object { payload }

payload: string

workers_ai_billing_mode: optional "postpaid" or "unified"

Controls how Workers AI inference calls routed through this gateway are billed. ‘postpaid’ bills the account directly through Workers AI; ‘unified’ deducts credits via AI Gateway using neuron-based pricing and delegates billing to AI Gateway.

One of the following:

"postpaid"

"unified"

zdr: optional boolean

AIGatewayGetResponse object { id, cache_invalidate_on_update, cache_ttl, 27 more }

id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

cache_invalidate_on_update: boolean

cache_ttl: number

minimum0

collect_logs: boolean

created_at: string

formatdate-time

modified_at: string

formatdate-time

rate_limiting_interval: number

minimum0

rate_limiting_limit: number

minimum0

authentication: optional boolean

Require authentication for gateway access. New gateways must have authentication enabled. Once enabled, authentication cannot be disabled.

byok_only: optional boolean

Requires customer-provided provider credentials and prevents fallback to Unified Billing.

dlp: optional object { action, enabled, profiles } or object { enabled, policies }

One of the following:

object { action, enabled, profiles }

action: "BLOCK" or "FLAG"

One of the following:

"BLOCK"

"FLAG"

enabled: boolean

profiles: array of string

object { enabled, policies }

enabled: boolean

policies: array of object { id, action, check, 2 more }

id: string

action: "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

check: array of "REQUEST" or "RESPONSE"

One of the following:

"REQUEST"

"RESPONSE"

enabled: boolean

profiles: array of string

fedramp_level: optional "fedramp-medium" or "fedramp-high"

FedRAMP compliance level: fedramp-medium or fedramp-high.

One of the following:

"fedramp-medium"

"fedramp-high"

fedramp_mode: optional "prefer" or "require"

FedRAMP enforcement mode. “prefer” prioritizes compliant connectors but allows fallback. “require” strictly filters.

One of the following:

"prefer"

"require"

guardrails: optional object { prompt, response }

prompt: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

response: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

is_default: optional boolean

log_classification: optional boolean

log_management: optional number

maximum10000000

minimum10000

log_management_strategy: optional "STOP_INSERTING" or "DELETE_OLDEST"

One of the following:

"STOP_INSERTING"

"DELETE_OLDEST"

logpush: optional boolean

logpush_public_key: optional string

maxLength1024

minLength16

otel: optional array of object { headers, url, authorization, content_type }

headers: map[string]

url: string

formaturi

maxLength2048

authorization: optional string

maxLength256

content_type: optional "json" or "protobuf"

One of the following:

"json"

"protobuf"

rate_limiting_technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

retry_backoff: optional "constant" or "linear" or "exponential"

Backoff strategy for retry delays

One of the following:

"constant"

"linear"

"exponential"

retry_delay: optional number

Delay between retry attempts in milliseconds (0-60000)

maximum60000

minimum0

retry_max_attempts: optional number

Maximum number of retry attempts for failed requests (1-5)

maximum5

minimum1

spend_limits: optional object { enabled, rules }

enabled: optional boolean

rules: optional array of object { limit, limitType, window, 6 more }

limit: number

exclusiveMinimum

minimum0

limitType: "cost"

window: number

exclusiveMinimum

minimum0

id: optional string

minLength1

enabled: optional boolean

metadata: optional map[object { mode } or object { mode, values } ]

One of the following:

Mode object { mode }

mode: "partition"

object { mode, values }

mode: "filter"

values: array of string

model: optional object { mode, values }

mode: "filter"

values: array of string

provider: optional object { mode, values }

mode: "filter"

values: array of string

technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

store_id: optional string

Secrets Store used by this gateway. Omit on update to keep the current store. Cannot be cleared while stored keys or secret references depend on it.

stripe: optional object { authorization, usage_events }

authorization: string

usage_events: array of object { payload }

payload: string

workers_ai_billing_mode: optional "postpaid" or "unified"

One of the following:

"postpaid"

"unified"

zdr: optional boolean

AIGatewayCreateResponse object { id, cache_invalidate_on_update, cache_ttl, 27 more }

id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

cache_invalidate_on_update: boolean

cache_ttl: number

minimum0

collect_logs: boolean

created_at: string

formatdate-time

modified_at: string

formatdate-time

rate_limiting_interval: number

minimum0

rate_limiting_limit: number

minimum0

authentication: optional boolean

Require authentication for gateway access. New gateways must have authentication enabled. Once enabled, authentication cannot be disabled.

byok_only: optional boolean

Requires customer-provided provider credentials and prevents fallback to Unified Billing.

dlp: optional object { action, enabled, profiles } or object { enabled, policies }

One of the following:

object { action, enabled, profiles }

action: "BLOCK" or "FLAG"

One of the following:

"BLOCK"

"FLAG"

enabled: boolean

profiles: array of string

object { enabled, policies }

enabled: boolean

policies: array of object { id, action, check, 2 more }

id: string

action: "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

check: array of "REQUEST" or "RESPONSE"

One of the following:

"REQUEST"

"RESPONSE"

enabled: boolean

profiles: array of string

fedramp_level: optional "fedramp-medium" or "fedramp-high"

FedRAMP compliance level: fedramp-medium or fedramp-high.

One of the following:

"fedramp-medium"

"fedramp-high"

fedramp_mode: optional "prefer" or "require"

FedRAMP enforcement mode. “prefer” prioritizes compliant connectors but allows fallback. “require” strictly filters.

One of the following:

"prefer"

"require"

guardrails: optional object { prompt, response }

prompt: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

response: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

is_default: optional boolean

log_classification: optional boolean

log_management: optional number

maximum10000000

minimum10000

log_management_strategy: optional "STOP_INSERTING" or "DELETE_OLDEST"

One of the following:

"STOP_INSERTING"

"DELETE_OLDEST"

logpush: optional boolean

logpush_public_key: optional string

maxLength1024

minLength16

otel: optional array of object { headers, url, authorization, content_type }

headers: map[string]

url: string

formaturi

maxLength2048

authorization: optional string

maxLength256

content_type: optional "json" or "protobuf"

One of the following:

"json"

"protobuf"

rate_limiting_technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

retry_backoff: optional "constant" or "linear" or "exponential"

Backoff strategy for retry delays

One of the following:

"constant"

"linear"

"exponential"

retry_delay: optional number

Delay between retry attempts in milliseconds (0-60000)

maximum60000

minimum0

retry_max_attempts: optional number

Maximum number of retry attempts for failed requests (1-5)

maximum5

minimum1

spend_limits: optional object { enabled, rules }

enabled: optional boolean

rules: optional array of object { limit, limitType, window, 6 more }

limit: number

exclusiveMinimum

minimum0

limitType: "cost"

window: number

exclusiveMinimum

minimum0

id: optional string

minLength1

enabled: optional boolean

metadata: optional map[object { mode } or object { mode, values } ]

One of the following:

Mode object { mode }

mode: "partition"

object { mode, values }

mode: "filter"

values: array of string

model: optional object { mode, values }

mode: "filter"

values: array of string

provider: optional object { mode, values }

mode: "filter"

values: array of string

technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

store_id: optional string

Secrets Store used by this gateway. Omit on update to keep the current store. Cannot be cleared while stored keys or secret references depend on it.

stripe: optional object { authorization, usage_events }

authorization: string

usage_events: array of object { payload }

payload: string

workers_ai_billing_mode: optional "postpaid" or "unified"

One of the following:

"postpaid"

"unified"

zdr: optional boolean

AIGatewayUpdateResponse object { id, cache_invalidate_on_update, cache_ttl, 27 more }

id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

cache_invalidate_on_update: boolean

cache_ttl: number

minimum0

collect_logs: boolean

created_at: string

formatdate-time

modified_at: string

formatdate-time

rate_limiting_interval: number

minimum0

rate_limiting_limit: number

minimum0

authentication: optional boolean

Require authentication for gateway access. New gateways must have authentication enabled. Once enabled, authentication cannot be disabled.

byok_only: optional boolean

Requires customer-provided provider credentials and prevents fallback to Unified Billing.

dlp: optional object { action, enabled, profiles } or object { enabled, policies }

One of the following:

object { action, enabled, profiles }

action: "BLOCK" or "FLAG"

One of the following:

"BLOCK"

"FLAG"

enabled: boolean

profiles: array of string

object { enabled, policies }

enabled: boolean

policies: array of object { id, action, check, 2 more }

id: string

action: "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

check: array of "REQUEST" or "RESPONSE"

One of the following:

"REQUEST"

"RESPONSE"

enabled: boolean

profiles: array of string

fedramp_level: optional "fedramp-medium" or "fedramp-high"

FedRAMP compliance level: fedramp-medium or fedramp-high.

One of the following:

"fedramp-medium"

"fedramp-high"

fedramp_mode: optional "prefer" or "require"

FedRAMP enforcement mode. “prefer” prioritizes compliant connectors but allows fallback. “require” strictly filters.

One of the following:

"prefer"

"require"

guardrails: optional object { prompt, response }

prompt: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

response: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

is_default: optional boolean

log_classification: optional boolean

log_management: optional number

maximum10000000

minimum10000

log_management_strategy: optional "STOP_INSERTING" or "DELETE_OLDEST"

One of the following:

"STOP_INSERTING"

"DELETE_OLDEST"

logpush: optional boolean

logpush_public_key: optional string

maxLength1024

minLength16

otel: optional array of object { headers, url, authorization, content_type }

headers: map[string]

url: string

formaturi

maxLength2048

authorization: optional string

maxLength256

content_type: optional "json" or "protobuf"

One of the following:

"json"

"protobuf"

rate_limiting_technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

retry_backoff: optional "constant" or "linear" or "exponential"

Backoff strategy for retry delays

One of the following:

"constant"

"linear"

"exponential"

retry_delay: optional number

Delay between retry attempts in milliseconds (0-60000)

maximum60000

minimum0

retry_max_attempts: optional number

Maximum number of retry attempts for failed requests (1-5)

maximum5

minimum1

spend_limits: optional object { enabled, rules }

enabled: optional boolean

rules: optional array of object { limit, limitType, window, 6 more }

limit: number

exclusiveMinimum

minimum0

limitType: "cost"

window: number

exclusiveMinimum

minimum0

id: optional string

minLength1

enabled: optional boolean

metadata: optional map[object { mode } or object { mode, values } ]

One of the following:

Mode object { mode }

mode: "partition"

object { mode, values }

mode: "filter"

values: array of string

model: optional object { mode, values }

mode: "filter"

values: array of string

provider: optional object { mode, values }

mode: "filter"

values: array of string

technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

store_id: optional string

Secrets Store used by this gateway. Omit on update to keep the current store. Cannot be cleared while stored keys or secret references depend on it.

stripe: optional object { authorization, usage_events }

authorization: string

usage_events: array of object { payload }

payload: string

workers_ai_billing_mode: optional "postpaid" or "unified"

One of the following:

"postpaid"

"unified"

zdr: optional boolean

AIGatewayDeleteResponse object { id, cache_invalidate_on_update, cache_ttl, 27 more }

id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

cache_invalidate_on_update: boolean

cache_ttl: number

minimum0

collect_logs: boolean

created_at: string

formatdate-time

modified_at: string

formatdate-time

rate_limiting_interval: number

minimum0

rate_limiting_limit: number

minimum0

authentication: optional boolean

Require authentication for gateway access. New gateways must have authentication enabled. Once enabled, authentication cannot be disabled.

byok_only: optional boolean

Requires customer-provided provider credentials and prevents fallback to Unified Billing.

dlp: optional object { action, enabled, profiles } or object { enabled, policies }

One of the following:

object { action, enabled, profiles }

action: "BLOCK" or "FLAG"

One of the following:

"BLOCK"

"FLAG"

enabled: boolean

profiles: array of string

object { enabled, policies }

enabled: boolean

policies: array of object { id, action, check, 2 more }

id: string

action: "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

check: array of "REQUEST" or "RESPONSE"

One of the following:

"REQUEST"

"RESPONSE"

enabled: boolean

profiles: array of string

fedramp_level: optional "fedramp-medium" or "fedramp-high"

FedRAMP compliance level: fedramp-medium or fedramp-high.

One of the following:

"fedramp-medium"

"fedramp-high"

fedramp_mode: optional "prefer" or "require"

FedRAMP enforcement mode. “prefer” prioritizes compliant connectors but allows fallback. “require” strictly filters.

One of the following:

"prefer"

"require"

guardrails: optional object { prompt, response }

prompt: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

response: object { P1, S1, S10, 12 more }

P1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S1: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S10: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S11: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S12: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S13: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S14: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S2: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S3: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S4: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S5: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S6: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S7: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S8: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

S9: optional "FLAG" or "BLOCK"

One of the following:

"FLAG"

"BLOCK"

is_default: optional boolean

log_classification: optional boolean

log_management: optional number

maximum10000000

minimum10000

log_management_strategy: optional "STOP_INSERTING" or "DELETE_OLDEST"

One of the following:

"STOP_INSERTING"

"DELETE_OLDEST"

logpush: optional boolean

logpush_public_key: optional string

maxLength1024

minLength16

otel: optional array of object { headers, url, authorization, content_type }

headers: map[string]

url: string

formaturi

maxLength2048

authorization: optional string

maxLength256

content_type: optional "json" or "protobuf"

One of the following:

"json"

"protobuf"

rate_limiting_technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

retry_backoff: optional "constant" or "linear" or "exponential"

Backoff strategy for retry delays

One of the following:

"constant"

"linear"

"exponential"

retry_delay: optional number

Delay between retry attempts in milliseconds (0-60000)

maximum60000

minimum0

retry_max_attempts: optional number

Maximum number of retry attempts for failed requests (1-5)

maximum5

minimum1

spend_limits: optional object { enabled, rules }

enabled: optional boolean

rules: optional array of object { limit, limitType, window, 6 more }

limit: number

exclusiveMinimum

minimum0

limitType: "cost"

window: number

exclusiveMinimum

minimum0

id: optional string

minLength1

enabled: optional boolean

metadata: optional map[object { mode } or object { mode, values } ]

One of the following:

Mode object { mode }

mode: "partition"

object { mode, values }

mode: "filter"

values: array of string

model: optional object { mode, values }

mode: "filter"

values: array of string

provider: optional object { mode, values }

mode: "filter"

values: array of string

technique: optional "fixed" or "sliding"

One of the following:

"fixed"

"sliding"

store_id: optional string

Secrets Store used by this gateway. Omit on update to keep the current store. Cannot be cleared while stored keys or secret references depend on it.

stripe: optional object { authorization, usage_events }

authorization: string

usage_events: array of object { payload }

payload: string

workers_ai_billing_mode: optional "postpaid" or "unified"

One of the following:

"postpaid"

"unified"

zdr: optional boolean

AI GatewayEvaluation Types

List evaluator types (deprecated)

GET/accounts/{account_id}/ai-gateway/evaluation-types

ModelsExpand Collapse

EvaluationTypeListResponse object { id, created_at, description, 5 more }

id: string

created_at: string

formatdate-time

description: string

enable: boolean

mandatory: boolean

modified_at: string

formatdate-time

type: string

AI GatewayCustom Providers

List custom providers

GET/accounts/{account_id}/ai-gateway/custom-providers

Get a custom provider

GET/accounts/{account_id}/ai-gateway/custom-providers/{id}

Create a custom provider

POST/accounts/{account_id}/ai-gateway/custom-providers

Delete a custom provider

DELETE/accounts/{account_id}/ai-gateway/custom-providers/{id}

ModelsExpand Collapse

CustomProviderListResponse object { id, base_url, created_at, 12 more }

id: string

formatuuid

base_url: string

formaturi

created_at: string

formatdate-time

modified_at: string

formatdate-time

slug: string

beta: optional boolean

curl_example: optional string

description: optional string

enable: optional boolean

headers: optional string

JSON object of extra HTTP headers that AI Gateway sends to the provider. Values can contain credentials.

maxLength8192

js_example: optional string

link: optional string

logo: optional string

position: optional number

CustomProviderGetResponse object { id, base_url, created_at, 12 more }

id: string

formatuuid

base_url: string

formaturi

created_at: string

formatdate-time

modified_at: string

formatdate-time

slug: string

beta: optional boolean

curl_example: optional string

description: optional string

enable: optional boolean

headers: optional string

JSON object of extra HTTP headers that AI Gateway sends to the provider. Values can contain credentials.

maxLength8192

js_example: optional string

link: optional string

logo: optional string

position: optional number

CustomProviderCreateResponse object { id, base_url, created_at, 12 more }

id: string

formatuuid

base_url: string

formaturi

created_at: string

formatdate-time

modified_at: string

formatdate-time

slug: string

beta: optional boolean

curl_example: optional string

description: optional string

enable: optional boolean

headers: optional string

JSON object of extra HTTP headers that AI Gateway sends to the provider. Values can contain credentials.

maxLength8192

js_example: optional string

link: optional string

logo: optional string

position: optional number

CustomProviderDeleteResponse object { id, base_url, created_at, 12 more }

id: string

formatuuid

base_url: string

formaturi

created_at: string

formatdate-time

modified_at: string

formatdate-time

slug: string

beta: optional boolean

curl_example: optional string

description: optional string

enable: optional boolean

headers: optional string

JSON object of extra HTTP headers that AI Gateway sends to the provider. Values can contain credentials.

maxLength8192

js_example: optional string

link: optional string

logo: optional string

position: optional number

AI GatewayLogs

List Gateway Logs

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/logs

Get Gateway Log Detail

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/logs/{id}

Patch Gateway Log

PATCH/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/logs/{id}

Delete Gateway Logs

DELETE/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/logs

Get Gateway Log Request

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/logs/{id}/request

Get Gateway Log Response

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/logs/{id}/response

ModelsExpand Collapse

LogListResponse object { id, cached, created_at, 17 more }

id: string

cached: boolean

created_at: string

formatdate-time

duration: number

model: string

path: string

provider: string

success: boolean

tokens_in: number

tokens_out: number

collect_log_payload: optional boolean

Whether the request and response bodies were eligible to be stored. False when the request sent cf-aig-collect-log-payload: false, logging was off, or the gateway does not store bodies for the request. Null for logs written before this field existed.

cost: optional number

custom_cost: optional boolean

metadata: optional string

model_type: optional string

request_content_type: optional string

request_type: optional string

response_content_type: optional string

status_code: optional number

step: optional number

LogGetResponse object { id, cached, created_at, 23 more }

id: string

cached: boolean

created_at: string

formatdate-time

duration: number

model: string

path: string

provider: string

success: boolean

tokens_in: number

tokens_out: number

collect_log_payload: optional boolean

cost: optional number

custom_cost: optional boolean

metadata: optional string

model_type: optional string

request_content_type: optional string

request_head: optional string

request_head_complete: optional boolean

request_size: optional number

request_type: optional string

response_content_type: optional string

response_head: optional string

response_head_complete: optional boolean

response_size: optional number

status_code: optional number

step: optional number

LogEditResponse = unknown

LogDeleteResponse object { success }

success: boolean

LogRequestResponse = unknown

LogResponseResponse = unknown

AI GatewayDatasets

List datasets (deprecated)

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/datasets

Get a dataset (deprecated)

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/datasets/{id}

Create a dataset (deprecated)

POST/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/datasets

Update a dataset (deprecated)

PUT/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/datasets/{id}

Delete a dataset (deprecated)

DELETE/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/datasets/{id}

ModelsExpand Collapse

DatasetListResponse object { id, created_at, enable, 4 more }

id: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

DatasetGetResponse object { id, created_at, enable, 4 more }

id: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

DatasetCreateResponse object { id, created_at, enable, 4 more }

id: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

DatasetUpdateResponse object { id, created_at, enable, 4 more }

id: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

DatasetDeleteResponse object { id, created_at, enable, 4 more }

id: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

AI GatewayEvaluations

List evaluations (deprecated)

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/evaluations

Get an evaluation (deprecated)

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/evaluations/{id}

Create an evaluation (deprecated)

POST/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/evaluations

Delete an evaluation (deprecated)

DELETE/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/evaluations/{id}

ModelsExpand Collapse

EvaluationListResponse object { id, created_at, datasets, 6 more }

id: string

created_at: string

formatdate-time

datasets: array of object { id, account_id, account_tag, 6 more }

id: string

account_id: string

account_tag: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

processed: boolean

results: array of object { id, created_at, evaluation_id, 6 more }

id: string

created_at: string

formatdate-time

evaluation_id: string

evaluation_type_id: string

modified_at: string

formatdate-time

result: string

status: number

status_description: string

total_logs: number

EvaluationGetResponse object { id, created_at, datasets, 6 more }

id: string

created_at: string

formatdate-time

datasets: array of object { id, account_id, account_tag, 6 more }

id: string

account_id: string

account_tag: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

processed: boolean

results: array of object { id, created_at, evaluation_id, 6 more }

id: string

created_at: string

formatdate-time

evaluation_id: string

evaluation_type_id: string

modified_at: string

formatdate-time

result: string

status: number

status_description: string

total_logs: number

EvaluationCreateResponse object { id, created_at, datasets, 6 more }

id: string

created_at: string

formatdate-time

datasets: array of object { id, account_id, account_tag, 6 more }

id: string

account_id: string

account_tag: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

processed: boolean

results: array of object { id, created_at, evaluation_id, 6 more }

id: string

created_at: string

formatdate-time

evaluation_id: string

evaluation_type_id: string

modified_at: string

formatdate-time

result: string

status: number

status_description: string

total_logs: number

EvaluationDeleteResponse object { id, created_at, datasets, 6 more }

id: string

created_at: string

formatdate-time

datasets: array of object { id, account_id, account_tag, 6 more }

id: string

account_id: string

account_tag: string

created_at: string

formatdate-time

enable: boolean

filters: array of object { key, operator, value }

key: "created_at" or "request_content_type" or "response_content_type" or 10 more

One of the following:

"created_at"

"request_content_type"

"response_content_type"

"success"

"cached"

"provider"

"model"

"cost"

"tokens"

"tokens_in"

"tokens_out"

"duration"

"feedback"

operator: "eq" or "contains" or "lt" or "gt"

One of the following:

"eq"

"contains"

"lt"

"gt"

value: array of string or number or boolean

One of the following:

string

number

boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

processed: boolean

results: array of object { id, created_at, evaluation_id, 6 more }

id: string

created_at: string

formatdate-time

evaluation_id: string

evaluation_type_id: string

modified_at: string

formatdate-time

result: string

status: number

status_description: string

total_logs: number

AI GatewayDynamic Routing

List dynamic routes

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes

Get a dynamic route

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}

Create a dynamic route

POST/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes

Rename a dynamic route

PATCH/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}

Delete a dynamic route

DELETE/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}

List dynamic route deployments

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}/deployments

Deploy a dynamic route version

POST/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}/deployments

List dynamic route versions

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}/versions

Create a dynamic route version

POST/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}/versions

Get a dynamic route version

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/routes/{id}/versions/{version_id}

ModelsExpand Collapse

DynamicRoutingListResponse object { data, success }

data: object { order_by, order_by_direction, page, 2 more }

order_by: string

order_by_direction: string

page: number

per_page: number

routes: array of object { id, account_tag, created_at, 6 more }

id: string

account_tag: string

created_at: string

formatdate-time

deployment: object { created_at, deployment_id, version_id }

created_at: string

deployment_id: string

version_id: string

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

version: object { active, created_at, data, 2 more }

active: "true" or "false"

One of the following:

"true"

"false"

created_at: string

data: string

version_id: string

is_valid: optional boolean

success: boolean

DynamicRoutingGetResponse object { id, created_at, deployment, 5 more }

id: string

created_at: string

formatdate-time

deployment: object { created_at, deployment_id, version_id }

created_at: string

deployment_id: string

version_id: string

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

version: object { active, created_at, data, 2 more }

active: "true" or "false"

One of the following:

"true"

"false"

created_at: string

data: string

version_id: string

is_valid: optional boolean

DynamicRoutingCreateResponse object { id, created_at, deployment, 5 more }

id: string

created_at: string

formatdate-time

deployment: object { created_at, deployment_id, version_id }

created_at: string

deployment_id: string

version_id: string

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

version: object { active, created_at, data, 2 more }

active: "true" or "false"

One of the following:

"true"

"false"

created_at: string

data: string

version_id: string

is_valid: optional boolean

DynamicRoutingUpdateResponse object { route, success }

route: object { id, account_tag, created_at, 6 more }

id: string

account_tag: string

created_at: string

formatdate-time

deployment: object { created_at, deployment_id, version_id }

created_at: string

deployment_id: string

version_id: string

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

version: object { active, created_at, data, 2 more }

active: "true" or "false"

One of the following:

"true"

"false"

created_at: string

data: string

version_id: string

is_valid: optional boolean

success: boolean

DynamicRoutingDeleteResponse object { id, created_at, elements, 3 more }

id: string

created_at: string

formatdate-time

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

DynamicRoutingListDeploymentsResponse object { data, success }

data: object { deployments, order_by, order_by_direction, 2 more }

deployments: array of object { created_at, deployment_id, version_id }

created_at: string

deployment_id: string

version_id: string

order_by: string

order_by_direction: string

page: number

per_page: number

success: boolean

DynamicRoutingCreateDeploymentResponse object { id, created_at, elements, 3 more }

id: string

created_at: string

formatdate-time

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

DynamicRoutingListVersionsResponse object { data, success }

data: object { order_by, order_by_direction, page, 2 more }

order_by: string

order_by_direction: string

page: number

per_page: number

versions: array of object { active, created_at, data, 2 more }

active: "true" or "false"

One of the following:

"true"

"false"

created_at: string

data: string

version_id: string

is_valid: optional boolean

success: boolean

DynamicRoutingCreateVersionResponse object { id, created_at, elements, 3 more }

id: string

created_at: string

formatdate-time

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

DynamicRoutingGetVersionResponse object { id, active, created_at, 7 more }

id: string

active: "true" or "false"

One of the following:

"true"

"false"

created_at: string

data: string

elements: array of object { id, outputs, type } or object { id, outputs, properties, type } or object { id, outputs, type } or 3 more

One of the following:

object { id, outputs, type }

id: string

outputs: object { next }

next: object { elementId }

elementId: string

type: "start"

object { id, outputs, properties, type }

id: string

outputs: object { false, true }

false: object { elementId }

elementId: string

true: object { elementId }

elementId: string

properties: object { conditions }

conditions: optional unknown

type: "conditional"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "percentage"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { key, limit, limitType, window }

key: string

limit: number

limitType: "count" or "cost"

One of the following:

"count"

"cost"

window: number

type: "rate"

object { id, outputs, properties, type }

id: string

outputs: object { fallback, success }

fallback: object { elementId }

elementId: string

success: object { elementId }

elementId: string

properties: object { model, provider, retries, timeout }

model: string

provider: string

retries: number

timeout: number

type: "model"

object { id, outputs, type }

id: string

outputs: map[object { elementId } ]

elementId: string

type: "end"

gateway_id: string

modified_at: string

formatdate-time

minLength1

version_id: string

is_valid: optional boolean

AI GatewayProvider Configs

List provider keys

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/provider_configs

Store a provider key

POST/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/provider_configs

ModelsExpand Collapse

ProviderConfigListResponse object { id, alias, default_config, 8 more }

id: string

alias: string

default_config: boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

provider_slug: string

secret_id: string

secret_preview: string

last_used_at: optional string

UTC date (YYYY-MM-DD) the gateway last selected this key to authenticate a request to the provider, whether or not the request succeeded. Updated at most once a day. Null if the gateway has not used the key since this field was introduced.

formatdate

rate_limit: optional number

rate_limit_period: optional number

ProviderConfigCreateResponse object { id, alias, default_config, 8 more }

id: string

alias: string

default_config: boolean

gateway_id: string

Unique identifier of the AI Gateway within the account.

maxLength64

minLength1

modified_at: string

formatdate-time

provider_slug: string

secret_id: string

secret_preview: string

last_used_at: optional string

formatdate

rate_limit: optional number

rate_limit_period: optional number

AI GatewayURLs

Get Gateway URL

GET/accounts/{account_id}/ai-gateway/gateways/{gateway_id}/url/{provider}

ModelsExpand Collapse

URLGetResponse = string

AI GatewayBilling

Get credit balance

GET/accounts/{account_id}/ai-gateway/billing/credit-balance

Get usage history

GET/accounts/{account_id}/ai-gateway/billing/usage-history

Get invoice history

GET/accounts/{account_id}/ai-gateway/billing/invoice-history

Get invoice preview

GET/accounts/{account_id}/ai-gateway/billing/invoice-preview

ModelsExpand Collapse

BillingCreditBalanceResponse object { balance, has_default_payment_method, payment_method, 2 more }

balance: number

has_default_payment_method: boolean

payment_method: object { brand, last4 }

brand: optional string

last4: optional string

topup_config: object { amount, threshold, disabledReason, 2 more }

amount: number

threshold: number

disabledReason: optional string

error: optional string

lastFailedAt: optional number

first_topup_success: optional boolean

BillingUsageHistoryResponse object { history }

history: array of object { id, aggregated_value, end_time, start_time }

id: string

aggregated_value: number

AI Gateway usage cost in the window, in US dollars.

end_time: number

start_time: number

BillingInvoiceHistoryResponse object { invoices, pagination }

invoices: array of object { amount_due, amount_paid, amount_remaining, 11 more }

amount_due: number

amount_paid: number

amount_remaining: number

currency: string

id: optional string

attempt_count: optional number

attempted: optional boolean

auto_advance: optional boolean

created: optional number

created_by: optional string

description: optional string

invoice_origin: optional string

invoice_pdf: optional string

status: optional string

pagination: object { has_more, page, per_page, total_count }

has_more: boolean

page: number

per_page: number

total_count: number

BillingInvoicePreviewResponse object { id, amount_due, amount_paid, 6 more }

id: string

amount_due: number

amount_paid: number

amount_remaining: number

currency: string

invoice_lines: array of object { amount, currency, description, 4 more }

amount: number

currency: string

description: string

period: object { end, start }

end: number

start: number

pricing: object { unit_amount_decimal }

unit_amount_decimal: string

quantity: number

pretax_credit_amounts: optional array of object { amount, type, credit_balance_transaction, discount }

amount: number

type: string

credit_balance_transaction: optional string

discount: optional string

period_end: number

period_start: number

status: "draft" or "open" or "paid" or 2 more

One of the following:

"draft"

"open"

"paid"

"uncollectible"

"void"

AI GatewayBillingTopup

Create a top-up

POST/accounts/{account_id}/ai-gateway/billing/topup

Check top-up status

POST/accounts/{account_id}/ai-gateway/billing/topup/status

ModelsExpand Collapse

TopupCreateResponse object { client_secret, onboarding, payment_intent_id, 2 more }

client_secret: string

Client secret used to complete payment authentication.

onboarding: boolean

Whether the user was already onboarded.

payment_intent_id: string

Top-up payment identifier.

brand: optional string

Card brand (visa, mastercard, etc.).

last4: optional string

Last 4 digits of card.

TopupStatusResponse object { payment_intent_id, status }

payment_intent_id: string

status: "completed" or "pending"

One of the following:

"completed"

"pending"

ModelsExpand Collapse

ConfigGetResponse object { amount, threshold, disabledReason, 2 more }

amount: number

threshold: number

disabledReason: optional string

error: optional string

lastFailedAt: optional number

ConfigCreateResponse = object { amount, threshold } or object { amount, status, threshold } or object { amount, client_secret, setup_intent_id, 2 more }

One of the following:

object { amount, threshold }

amount: number

threshold: number

object { amount, status, threshold }

amount: number

status: "configured"

threshold: number

object { amount, client_secret, setup_intent_id, 2 more }

amount: number

client_secret: string

setup_intent_id: string

status: "requires_action"

threshold: number

ConfigDeleteResponse = unknown

ModelsExpand Collapse

SpendingLimitGetResponse object { config, enabled }

config: object { amount, duration, strategy }

amount: number

duration: string

strategy: string

enabled: boolean

SpendingLimitCreateResponse = unknown

SpendingLimitDeleteResponse = unknown

AI Gateway

ModelsExpand Collapse

AI GatewayEvaluation Types

ModelsExpand Collapse

AI GatewayCustom Providers

ModelsExpand Collapse

AI GatewayLogs

ModelsExpand Collapse

AI GatewayDatasets

ModelsExpand Collapse

AI GatewayEvaluations

ModelsExpand Collapse

AI GatewayDynamic Routing

ModelsExpand Collapse

AI GatewayProvider Configs

ModelsExpand Collapse

AI GatewayURLs

ModelsExpand Collapse

AI GatewayBilling

ModelsExpand Collapse

AI GatewayBillingTopup

ModelsExpand Collapse

AI GatewayBillingTopupConfig

ModelsExpand Collapse

AI GatewayBillingSpending Limit

ModelsExpand Collapse